106 MASS. EXPERIMENT STATION BULLETIN 199. 



broody to 3 non-brood}^, exactly reversing the ratio, as has occurred in 

 the detailed data from which Table II was compiled. It should be stated 

 that while families showing the extreme ratio of 15 non-broody to 1 broody 

 have not been encountered, several instances of the 7:1 ratio have been 

 observed. 



The only evidence at present available in support of either of these 

 schemes is furnished bj^ the ratios between the broody and non-broody 

 members of the several famiUes (Table II). N, if it represents a real con- 

 dition of the germ plasm, occurs relatively infrequentlj'' in the flock at 

 present. Practically all the observed ratios, except the partial exception 

 mentioned, can be accounted for if N is omitted. 



It is also possible to modify the AACC theory, by assuming that A is 

 sex linked, though no evidence of sex linkage other than an agreement 

 between the observed and theoretical ratios has been noted. Doubtless 

 other schemes could be devised that would also account for the ratios. 



Although the ratios themselves could perhaps be explained as chance 

 deviations from monohybrid ratios (though this is doubtful in some 

 instances), or as the result of errors of classification of indi\'iduals through 

 failure to manifest the genotypic condition pha^notj'pically, the moment 

 lines of descent are established it becomes clear that a monoh3'brid ex- 

 planation does not fit the facts. The data have been worked over in 

 an attempt to apply the monohybrid scheme, i.e., broodiness due to a 

 single dominant factor, but without success. See, for example, the his- 

 tory of male No. 3003 and his offspring, page 107. 



In order to establish the existence of any of the schemes under dis- 

 cussion certain results of critical importance must be obtamed. Thus, 

 the discovery of a family consisting of all non-broody offspring from the 

 mating between a non-broody and a broody is required to demonstrate 

 the presence of a dominant factor for non-broodiness, while a mating 

 between two non-broody birds that gives all broodies is requked as proof 

 of the AACC theory (or a theory of the same order). The ratios at hand 

 indicate the possibility that several genetic types of non-broodies co- 

 exist in our strain. One possibilitj^ only seems to be excluded if the schemes 

 outlined represent the facts, for one need never expect to find a pair of 

 broody birds that produces all non-broody offspring, because such a result 

 would mean two distinct types of broodies which mutually inhibit each 

 other. 



Modifying Factors for Broodiness. 



The possibility that the non-broodies dealt with in these experiments 

 are not due to changes in the primary genes concerned with broodiness, 

 but are due to changes in modif3dng genes, cannot be excluded. As we 

 have worked over the records, the impression has been strong that we are 

 not dealing with a real absence of broodiness so much as with delay in 

 the appearance of broodiness. Unfortunately the present data are inade- 

 quate to settle this point. Noo: is it likelj' that we shall have suitable 

 data in the near future, because the somatic manifestations of broodiness, 



